Research project OR/10/006 (Research action OR)
Objectives
With the launch of PLEIADES in a near future, the availability of very high spatial resolution(VHR) satellite imagery will increase. These images are compatible with the large scale of base maps and offer a supporting potential to update these maps in quickly changing area or soon after natural hazards. Furthermore, the ability to combine information from a vector database and an image leads to enhanced performance for end users performing GIS analysis.
This project proposes a methodological approach mainly associated with the cartography of forest and natural vegetation. Its major outputs will be applicable to several themes, out of which “Forest”, “Agriculture” and “Cartography”.
This study focuses on an advanced method to complete, enrich and update a large scale vector database using VHR images. Existing vector information combined with VHR image- segments generates additional contextual and dynamic information beside the features obtained from image or vector separately. The originality of this research is to propose a robust classification scheme where spectrally irrelevant pixels are first concealed and then reintroduced as meaningful contextual information. It aims to implement an operational workflow to obtain a good positional and thematic accuracy using a single VHR image
Methodology
- Confrontation of the image and the vector database
The resolution of conflicts between the 2 databases is a prior step necessary for a good positional accuracy and a robust classification. First, a mathematical model will be used to locate areas with horizontal shift or shade based on the vector database and the image metadata. Second, boundaries will be adjusted using a combination of decision rule. Third, spectral artefacts in the resulting objects will be screen from representative pixels.
- Contextual assimilation of the image and the vector database.
Artefacts and representative pixels will be classified separately in an object based scheme. Specific object-based feature will be selected with a statistical approach. Both object sets will then be combined using decision rules and contextual information in order to be assimilated into the vector database.
Study area
Spa, Famenne and Montpellier
Data
IKONOS, Quickbird and simulated PLEIADES data
Existing vector database.
